The human immunodeficiency virus (HIV), a lentivirus, is an important target of anti-AIDS therapy. An essential step of the HIV lifecycle, and therefore a potential anti-HIV target, is transcriptional transactivation in which the HIV Tat protein recognizes the HIV TAR hairpin by weak but specific binding of its arginine-rich domain, with the participation of a cellular factor binding the TAR loop to stabilize the interaction. In contrast, bovine immunodeficiency virus (BIV) Tat recognizes BIV TAR with high affinity by its arginine-rich domain, with no requirement for cellular factors. Solution state modeling of HIV TAR-argininamide and BIV TAR-peptide complexes derived from NMR data indicate no obvious reason why BIV peptide cannot bind to HIV TAR with the same high affinity as to BIV TAR. Biochemical analysis indicates that HIV TAR is defective for BIV peptide binding because of sequences flanking the binding site. As a result of these studies, we have found TARs and peptides that can function with both HIV and BIV Tats and TARs, respectively. We are using these bifunctional molecules to explore the inhibition of lentiviral transactivation and to direct our search for peptides that will inhibit HIV transactivation and have anti-AIDS therapeutic potential. Our project relies upon visualization of NMR-derived and constructed models for rational design. The Computer Graphics Lab provides access to MidasPlus which we use to visualize existing models and help interpret how designed peptides bind to HIV TAR.